To set up the procedures for system management
System management strategy defined
1. Define the access procedures (firewalls, secure ID cards, ...) for remote support.
2. Define the start/stop procedures for the new solution. In distributed systems, you may need to stop and start the individual components in a special sequence.
3. Define a maintenance window. During that time, the system is not available for production use.
4. Define a backup strategy for the new solution. The backups should be consistent over all parts of the new solution. Check the readability of backup tapes at regular intervals.
5. Define disaster recovery plans. The plans must include the following topics:
- Possible errors (physical, hardware, or logical)
- For each error, the possible recovery scenarios
- Estimated time frame for each recovery scenario
- Scenarios that involve loss of data
- Point-in-time recovery (especially for distributed systems)
- Aspects not covered by the recovery plans (such as costs)
- Tests and documentation for these scenarios.
6. Define the required privileges for the administration of the new solution.
- Grant individual administrators only the rights they need to do their job (and not unlimited rights).
- Define an audit procedure for administrative actions and ensure that each major change in the system is documented.
- Use only named accounts for administrators (and not collective accounts like root in Unix systems).
- If possible, use a double-check strategy where one user makes changes and another user confirms the changes.
- Define a password policy for administrative accounts to specify how long a password is valid and how it must be changed.
System management procedures in place